pith. sign in

arxiv: 2605.21782 · v1 · pith:AVQSLVQRnew · submitted 2026-05-20 · 📊 stat.ME · stat.AP· stat.CO

A Scalable Parametric Item Calibration Engine (SPICE) for Explanatory IRT with Sparse Data

Steven W. Nydick , Manqian Liao , J.R. Lockwood This is my paper

Pith reviewed 2026-05-22 08:18 UTC · model grok-4.3

classification 📊 stat.ME stat.APstat.CO
keywords Bayesian IRTexplanatory item response theorysparse dataMCMC estimationitem calibrationadaptive assessmentmultidimensional IRTpsychometric software
0
0 comments X

The pith

A Bayesian MCMC procedure scales explanatory IRT calibration to large sparse datasets from adaptive testing.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents a software engine called SPICE that implements a Bayesian multidimensional explanatory item response theory model together with an MCMC estimation routine. The design targets psychometric data in which large numbers of persons and items connect only sparsely, the pattern that appears when adaptive tests draw from extensive automatically generated item banks and each test taker sees only a small fraction of the bank. Specific decisions about model form, data representation, and sampling algorithm are combined to reach computational scalability. A sympathetic reader would care because the method would make explanatory analyses feasible in settings where traditional dense-matrix approaches break down due to data sparsity.

Core claim

We describe a Bayesian multidimensional explanatory IRT model, and an associated Markov Chain Monte Carlo (MCMC) estimation procedure and the corresponding development of calibration software, designed for psychometric analyses of large numbers of sparsely-linked persons and items. Such data structures can arise, for example, from adaptive assessments using large banks of automatically generated items with individual test takers receiving a very small proportion of the entire bank. We discuss how our choices for model specification, data structures, and algorithm implementation combine to create a scalable method for explanatory IRT that can support a variety of psychometric operations with

What carries the argument

The SPICE calibration engine, which integrates a Bayesian multidimensional explanatory IRT model with an MCMC estimation procedure to perform scalable analysis on sparsely linked persons and items.

If this is right

  • Enables explanatory IRT analyses on large item banks where each person encounters only a small subset of items.
  • Supports a range of psychometric operations including calibration under sparse linking patterns.
  • Allows incorporation of explanatory variables for both items and persons without requiring dense response matrices.
  • Provides practical software for MCMC-based calibration in settings that generate automatically produced items.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same sparse-connection structure appears in recommendation systems and online learning platforms, so the modeling choices could transfer to those domains.
  • Performance on real operational logs from adaptive tests would supply a direct test of whether the scalability claims hold outside simulation.
  • Adding richer explanatory predictors or hierarchical structures could be examined while monitoring whether the MCMC implementation retains its speed advantage.

Load-bearing premise

That particular choices of model specification, data structures, and algorithm implementation will together produce a method that remains scalable for sparse data arising from adaptive assessments.

What would settle it

Apply the SPICE software to a large simulated or real adaptive-testing dataset containing thousands of items and persons with only sparse connections, then check whether the MCMC chains converge to stable estimates whose accuracy matches known values from denser data versions of the same items.

read the original abstract

We describe a Bayesian multidimensional explanatory IRT model, and an associated Markov Chain Monte Carlo (MCMC) estimation procedure and the corresponding development of calibration software, designed for psychometric analyses of large numbers of sparsely-linked persons and items. Such data structures can arise, for example, from adaptive assessments using large banks of automatically generated items with individual test takers receiving a very small proportion of the entire bank. We discuss how our choices for model specification, data structures, and algorithm implementation combine to create a scalable method for explanatory IRT that can support a variety of psychometric operations with sparse data.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript describes a Bayesian multidimensional explanatory Item Response Theory (IRT) model together with a Markov Chain Monte Carlo (MCMC) estimation procedure and the associated calibration software SPICE. The approach is intended for psychometric analyses involving large numbers of sparsely linked persons and items, such as those arising in adaptive assessments that draw from extensive banks of automatically generated items where each test-taker sees only a small fraction of the bank. The central claim is that the particular choices of model specification, data structures for sparse linking, and algorithm implementation together yield a scalable engine capable of supporting a range of explanatory IRT operations on such data.

Significance. If the scalability claim is substantiated with concrete evidence, the work could provide a practically useful addition to the toolkit for explanatory IRT in large-scale sparse settings common to modern adaptive testing. The explicit focus on combining modeling, data-structure, and implementation decisions to address sparsity is a constructive framing that could inform subsequent software development in the field.

major comments (1)
  1. [Abstract] Abstract and opening sections: the assertion that the chosen model specification, data structures, and MCMC implementation 'combine to create a scalable method' is presented as the load-bearing contribution, yet the manuscript supplies no performance metrics, convergence diagnostics, wall-clock timings, or scaling experiments that would demonstrate feasibility when item banks reach thousands of items and response density falls below 5%. Without such evidence the central claim remains unverified.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading and for identifying the need for concrete empirical support of the scalability claims. We address the comment below and describe the revisions that will be made.

read point-by-point responses

  1. Referee: [Abstract] Abstract and opening sections: the assertion that the chosen model specification, data structures, and MCMC implementation 'combine to create a scalable method' is presented as the load-bearing contribution, yet the manuscript supplies no performance metrics, convergence diagnostics, wall-clock timings, or scaling experiments that would demonstrate feasibility when item banks reach thousands of items and response density falls below 5%. Without such evidence the central claim remains unverified.

    Authors: We agree that the current manuscript does not contain explicit scaling experiments, wall-clock timings, or convergence diagnostics at the scale mentioned. The scalability argument in the paper rests on the combination of the sparse data structures, the explanatory model parameterization, and the tailored MCMC implementation that avoids dense matrix operations. To strengthen the central claim, the revised manuscript will add a dedicated empirical evaluation section. This will report wall-clock timings and effective sample sizes for item banks of 1,000–5,000 items at response densities of 1–5%, using both simulated data and a real large-scale assessment dataset. We will also include Gelman–Rubin statistics and trace-plot summaries for representative parameters. These additions will directly test feasibility under the conditions the referee identifies. revision: yes

Circularity Check

0 steps flagged

No circularity; new modeling and software approach is self-contained

full rationale

The paper presents a Bayesian multidimensional explanatory IRT model, MCMC estimation procedure, and calibration software explicitly designed for large sparse data from adaptive assessments. The central claim is that particular choices of model specification, data structures, and algorithm implementation combine to produce scalability, but this is advanced as an engineering and design outcome rather than any derivation that reduces by construction to fitted inputs, self-definitions, or prior self-citations. No equations equate predictions to parameters already estimated within the same system, and no uniqueness theorems or ansatzes are imported via self-citation chains. The work is therefore self-contained as a proposed method whose performance claims rest on implementation details and empirical testing outside the derivation itself.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review provides no explicit free parameters, axioms, or invented entities; the central claim rests on the untested assertion that the chosen model, data structures, and MCMC implementation will scale.

pith-pipeline@v0.9.0 · 5633 in / 1088 out tokens · 50664 ms · 2026-05-22T08:18:40.310588+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

150 extracted references · 150 canonical work pages · 3 internal anchors

  1. [1]

    Biometrika , author =

    Discontinuous. Biometrika , author =. 2020 , keywords =. doi:10.1093/biomet/asz083 , language =

    work page doi:10.1093/biomet/asz083 2020

  2. [2]

    2021 , eprint=

    Mixed Hamiltonian Monte Carlo for mixed discrete and continuous variables , author=. 2021 , eprint=

    work page 2021

  3. [3]

    Computer Methods in Applied Mechanics and Engineering , author =

    Accelerating. Computer Methods in Applied Mechanics and Engineering , author =. 2025 , keywords =. doi:10.1016/j.cma.2025.118401 , language =

    work page doi:10.1016/j.cma.2025.118401 2025

  4. [4]

    A Conceptual Introduction to Hamiltonian Monte Carlo

    Betancourt, Michael , month = jul, year =. A conceptual introduction to. doi:10.48550/arXiv.1701.02434 , urldate =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1701.02434

  5. [5]

    and Kramarz, F

    Abowd, J. and Kramarz, F. and Margolis, D. , title =. Econometrica , year = 1999, volume = 67, number = 2, pages =

    work page 1999

  6. [6]

    and Schank, T

    Andrews, M. and Schank, T. and Upward, R. , title =. The Stata Journal , year = 2006, volume = 6, number = 4, pages =

    work page 2006

  7. [7]

    2000 , journal =

    Barnard, John and McCulloch, Robert and Meng, Xiao-Li , title =. 2000 , journal =

    work page 2000

  8. [8]

    , title =

    Billingsley, P. , title =

    work page

  9. [9]

    and Tiao, G.C

    Box, G.E.P. and Tiao, G.C. , title =

    work page

  10. [10]

    Browne, W. J. and Goldstein, H. and Rasbash, J. , title =. 2001 , journal =

    work page 2001

  11. [11]

    Journal of the Royal Statistical Society, Series A: Statistics in Society , year = 1999, volume = 162, number =

    Clayton, David and Rasbash, Jon , title =. Journal of the Royal Statistical Society, Series A: Statistics in Society , year = 1999, volume = 162, number =

    work page 1999

  12. [12]

    , title =

    Gelman, A. , title =. Bayesian Analysis , year = 2006, volume = 1, number = 3, pages =

    work page 2006

  13. [13]

    and Xiao-Li Meng and H

    Gelman, A. and Xiao-Li Meng and H. Stern , title =. Statistica Sinica , year = 1996, volume = 6, pages =

    work page 1996

  14. [14]

    B., 1992, @doi [Statistical Science] 10.1214/ss/1177011136 , https://ui.adsabs.harvard.edu/abs/1992StaSc...7..457G 7, 457

    Inference from iterative simulation using multiple sequences , journal =. doi:10.1214/ss/1177011136 , author =

    work page doi:10.1214/ss/1177011136

  15. [15]

    , title =

    Gentle, J.E. , title =

    work page

  16. [16]

    , title =

    Goldstein, H. , title =. Sociological Methods and Research , year =

    work page

  17. [17]

    , title =

    Hanushek, E.A. , title =. The Journal of Human Resources , year =

    work page

  18. [18]

    Lindley, D. V. and Smith, A. F. M. , title =. 1972 , journal =

    work page 1972

  19. [19]

    and McCaffrey, D.F

    Lockwood, J.R. and McCaffrey, D.F. and Mariano, L.T. and Setodji, C. , title =. 2007 , journal =

    work page 2007

  20. [20]

    and McCaffrey, D.F

    Mariano, L.T. and McCaffrey, D.F. and Lockwood, J.R. , title =. Journal of Educational and Behavioral Statistics , year = 2010, volume = 35, number = 3, pages =

    work page 2010

  21. [21]

    doi:10.3102/10769986029001067 , author =

    Models for value-added modeling of teacher effects , journal =. doi:10.3102/10769986029001067 , author =

    work page doi:10.3102/10769986029001067

  22. [22]

    and Hanushek, Eric A

    Rivkin, Steven G. and Hanushek, Eric A. and Kain, John F. , title =. Econometrica , year = 2005, volume = 73, number = 2, pages =

    work page 2005

  23. [23]

    , title =

    Zellner, A. , title =. Journal of the American Statistical Association , year = 1962, volume = 57, number =

    work page 1962

  24. [24]

    Applied Psychological Measurement , volume=

    The multidimensional random coefficients multinomial logit model , author=. Applied Psychological Measurement , volume=. 1997 , publisher=. doi:10.1177/0146621697211001

    work page doi:10.1177/0146621697211001 1997

  25. [25]

    Equate: Observed-Score Linking and Equating , shorttitle =

    Albano, Anthony , year =. Equate: Observed-Score Linking and Equating , shorttitle =

    work page

  26. [26]

    and Chib, Siddhartha , title =

    Albert, James H. and Chib, Siddhartha , title =. Journal of the American Statistical Association , year = 1993, volume = 88, number = 422, pages =

    work page 1993

  27. [27]

    and Bai, Z

    Anderson, E. and Bai, Z. and Bischof, C. and Blackford, S. and Demmel, J. and Dongarra, J. and Du Croz, J. and Greenbaum, A. and Hammarling, S. and McKenney, A. and Sorensen, D. , year =

    work page

  28. [28]

    Kequate: The Kernel Method of Test Equating , shorttitle =

    Andersson, Bj. Kequate: The Kernel Method of Test Equating , shorttitle =. 2022 , month = apr, abstract =

    work page 2022

  29. [29]

    2017 , journal =

    Computational Psychometrics in Support of Collaborative Educational Assessments , author =. 2017 , journal =

    work page 2017

  30. [30]

    , title =

    Bejar, I.I. , title =. Test Theory for a New Generation of Tests , editors =

    work page

  31. [31]

    and Srinivasan, C

    Bhapkar, V.P. and Srinivasan, C. , title =. Annals of the Institute of Statistical Mathematics , volume =. 1994 , doi =

    work page 1994

  32. [32]

    ACM Transactions on Mathematical Software , author =

    An updated set of basic linear algebra subprograms (. ACM Transactions on Mathematical Software , author =. 2002 , keywords =. doi:10.1145/567806.567807 , language =

    work page doi:10.1145/567806.567807 2002

  33. [33]

    Bolsinova, Maria and Maris, Gunter , year =. Can. Frontiers in Psychology , volume =

    work page

  34. [34]

    2021 , pages=

    Journal of Statistical Software , author=. 2021 , pages=

    work page 2021

  35. [35]

    Applied Linguistics , note =

    Cai, Huiying and Yan, Xun and Chuang, Ping-Lin and Pan, Yulin and Huo, Mingyue , title =. Applied Linguistics , note =. 2025 , issn =

    work page 2025

  36. [36]

    Cardwell, Ramsey and LaFlair, Geoffrey T and Settles, Burr , year =

    work page

  37. [37]

    Psychometrika , author=

    Psychometrics Behind Computerized Adaptive Testing , volume=. Psychometrika , author=. 2015 , pages=. doi:10.1007/s11336-014-9401-5 , number=

    work page doi:10.1007/s11336-014-9401-5 2015

  38. [38]

    Measurement: Interdisciplinary Research and Perspectives , volume =

    Seungwon Chung and Carrie Houts , title =. Measurement: Interdisciplinary Research and Perspectives , volume =. 2020 , publisher =. doi:10.1080/15366367.2019.1693825 , url =

    work page doi:10.1080/15366367.2019.1693825 2020

  39. [39]

    Frontiers in Education , VOLUME=

    Circi, Ruhan and Hicks, Juanita and Sikali, Emmanuel , TITLE=. Frontiers in Education , VOLUME=. 2023 , DOI=

    work page 2023

  40. [40]

    Computerworld , author =

    When is it faster to have 64 bits? , url =. Computerworld , author =

    work page

  41. [41]

    , journal=

    Cohen, M. , journal=. The. 1968 , volume=

    work page 1968

  42. [42]

    and Thomas, Joy A

    Cover, Thomas M. and Thomas, Joy A. , year =. Elements of Information Theory , publisher =

    work page

  43. [43]

    OpenMP: an industry standard API for shared-memory programming , journal =

    Leonardo Dagum and Ramesh Menon , year =. OpenMP: an industry standard API for shared-memory programming , journal =

    work page

  44. [44]

    Psychometrika , pages =

    Random item. Psychometrika , pages =. 2008 , volume =

    work page 2008

  45. [45]

    2004 , publisher=

    Explanatory item response models: A generalized linear and nonlinear approach , author=. 2004 , publisher=

    work page 2004

  46. [46]

    Applied Psychological Measurement , volume=

    Improving the quality of ability estimates through multidimensional scoring and incorporation of ancillary variables , author=. Applied Psychological Measurement , volume=. 2009 , doi =

    work page 2009

  47. [47]

    Investigating Repeater Effects on Small Sample Equating: Include or Exclude? , shorttitle =

    Diao, Hongyu and Keller, Lisa , year =. Investigating Repeater Effects on Small Sample Equating: Include or Exclude? , shorttitle =. Applied Measurement in Education , volume =. doi:10.1080/08957347.2019.1674302 , langid =

    work page doi:10.1080/08957347.2019.1674302 2019

  48. [48]

    2004 , month = jul, journal =

    Equating, Concordance, and Expectation , author =. 2004 , month = jul, journal =. doi:10.1177/0146621604265031 , abstract =

    work page doi:10.1177/0146621604265031 2004

  49. [49]

    Seamless

    Eddelbuettel, Dirk , year =. Seamless

    work page

  50. [50]

    Fox, Jean-Paul , title =

    work page

  51. [51]

    Fox, Jean-Paul and Glas, Cees A. W. , title=. Psychometrika , year=. doi:10.1007/BF02294839 , number=

    work page doi:10.1007/bf02294839

  52. [52]

    Glas, Cees A. W. and. Computerized Adaptive Testing With Item Cloning , journal =. 2003 , doi =

    work page 2003

  53. [53]

    Goodwin, Sarah and Attali, Yigal and LaFlair, Geoffrey T and Park, Yena and Runge, Andrew , year =

    work page

  54. [54]

    Multiple Imputation of Multilevel Missing Data: An Introduction to the

    Grund, Simon and L. Multiple Imputation of Multilevel Missing Data: An Introduction to the. 2016 , month = oct, journal =. doi:10.1177/2158244016668220 , abstract =

    work page doi:10.1177/2158244016668220 2016

  55. [55]

    ACM Transactions on Mathematical Software , author =

    Rectangular full packed format for. ACM Transactions on Mathematical Software , author =. 2010 , keywords =. doi:10.1145/1731022.1731028 , abstract =

    work page doi:10.1145/1731022.1731028 2010

  56. [56]

    2021 , publisher=

    Computational psychometrics: New methodologies for a new generation of digital learning and assessment: With examples in. 2021 , publisher=

    work page 2021

  57. [57]

    Thomas Hofmann, Bernhard Sch¨ olkopf, and Alexander J Smola

    Adjustment by Minimum Discriminant Information , author =. 1984 , month = sep, journal =. doi:10.1214/aos/1176346715 , abstract =

    work page doi:10.1214/aos/1176346715 1984

  58. [58]

    and Wollang, F-J.M

    Hommel, B.E. and Wollang, F-J.M. and Kotova, V. and Zacher, H. and Schmukle, S.C. , title =. Psychometrika , volume =. 2022 , doi =

    work page 2022

  59. [59]

    Journal of Educational and Behavioral Statistics , volume =

    Sijia Huang and Li Cai , title =. Journal of Educational and Behavioral Statistics , volume =. 2024 , doi =

    work page 2024

  60. [60]

    Isbell and Benjamin Kremmel , title =

    Daniel R. Isbell and Benjamin Kremmel , title =. Language Testing , volume =. 2020 , doi =

    work page 2020

  61. [61]

    Jewsbury and Peter W

    Paul A. Jewsbury and Peter W. van Rijn , title =. Journal of Educational and Behavioral Statistics , volume =. 2020 , doi =

    work page 2020

  62. [62]

    and Jenkins, Frank , title =

    Johnson, Matthew S. and Jenkins, Frank , title =. ETS Research Report Series , volume =. doi:10.1002/j.2333-8504.2004.tb01965.x , year =

    work page doi:10.1002/j.2333-8504.2004.tb01965.x 2004

  63. [63]

    Johnson and Sandip Sinharay , title =

    Matthew S. Johnson and Sandip Sinharay , title =. Applied Psychological Measurement , volume =. 2005 , doi =

    work page 2005

  64. [64]

    2013 , keywords =

    fastlog , author =. 2013 , keywords =

    work page 2013

  65. [65]

    Journal of Educational Measurement , author =

    A comparative study of. Journal of Educational Measurement , author =. 2006 , keywords =. doi:10.1111/j.1745-3984.2006.00021.x , language =

    work page doi:10.1111/j.1745-3984.2006.00021.x 2006

  66. [66]

    and Song, D

    Multiple Imputation of Missing or Faulty Values Under Linear Constraints , author =. 2014 , month = jul, journal =. doi:10.1080/07350015.2014.885435 , abstract =

    work page doi:10.1080/07350015.2014.885435 2014

  67. [67]

    2014 , publisher =

    Test Equating, Scaling, and Linking , author =. 2014 , publisher =. doi:10.1007/978-1-4939-0317-7 , isbn =

    work page doi:10.1007/978-1-4939-0317-7 2014

  68. [68]

    An Optimized

    Christoph K\". An Optimized. Applied Psychological Measurement , volume =. 2020 , doi =

    work page 2020

  69. [69]

    Anderson, D.A

    On Information and Sufficiency , author =. 1951 , month = mar, journal =. doi:10.1214/aoms/1177729694 , abstract =

    work page doi:10.1214/aoms/1177729694 1951

  70. [70]

    and Staples, W

    Laduca, A. and Staples, W. I. and Templeton, B. and Holzman, G. B. , title =. Medical Education , volume =. doi:10.1111/j.1365-2923.1986.tb01042.x , url =. https://asmepublications.onlinelibrary.wiley.com/doi/pdf/10.1111/j.1365-2923.1986.tb01042.x , abstract =

    work page doi:10.1111/j.1365-2923.1986.tb01042.x 1986

  71. [71]

    LaFlair, Geoffrey T and Settles, Burr , year =

    work page

  72. [72]

    and Langenfeld, Thomas and Baig, Basim and Horie, Andr

    LaFlair, Geoffrey T. and Langenfeld, Thomas and Baig, Basim and Horie, Andr. Digital-First Assessments: A Security Framework , shorttitle =. 2022 , month = mar, journal =. doi:10.1111/jcal.12665 , abstract =

    work page doi:10.1111/jcal.12665 2022

  73. [73]

    2022 , month = may, journal =

    Digital-First Learning and Assessment Systems for the 21st Century , author =. 2022 , month = may, journal =. doi:10.3389/feduc.2022.857604 , abstract =

    work page doi:10.3389/feduc.2022.857604 2022

  74. [74]

    Levy, Roy and Mislevy, Robert J , year=

    work page

  75. [75]

    Generating random correlation matrices based on vines and extended onion method , journal =

    Daniel Lewandowski and Dorota Kurowicka and Harry Joe , keywords =. Generating random correlation matrices based on vines and extended onion method , journal =. 2009 , issn =. doi:https://doi.org/10.1016/j.jmva.2009.04.008 , url =

    work page doi:10.1016/j.jmva.2009.04.008 2009

  76. [76]

    Lissitz , title =

    Ming Li and Hong Jiao and Tianyi Zhou and Nan Zhang and Sydney Peters and Robert W. Lissitz , title =. Educational and Psychological Measurement , volume =. 2025 , doi =

    work page 2025

  77. [77]

    British Journal of Mathematical and Statistical Psychology , volume =

    Lin, Xiaofan and Zhang, Siliang and Tang, Yincai and Li, Xuan , title =. British Journal of Mathematical and Statistical Psychology , volume =. doi:https://doi.org/10.1111/bmsp.12321 , url =. https://bpspsychub.onlinelibrary.wiley.com/doi/pdf/10.1111/bmsp.12321 , abstract =

    work page doi:10.1111/bmsp.12321

  78. [78]

    Psychometrika , volume=

    Liu, Yang and Yang, Ji Seung , title=. Psychometrika , volume=. 2018 , pages=

    work page 2018

  79. [79]

    Journal of Educational and Behavioral Statistics , volume =

    Yang Liu and Ji Seung Yang , title =. Journal of Educational and Behavioral Statistics , volume =. 2018 , doi =

    work page 2018

  80. [80]

    Equating Test Scores (without

    Livingston, Samuel A , year =. Equating Test Scores (without

    work page

Showing first 80 references.